μSVM - A new method for solving the problem of imbalanced dataset classification

It is shown that SVM can be ineffective in classifying the minority sample, when it is applied to the problem of learning from imbalanced datasets. To remedy this problem, this paper analyzes the true cause of that problem firstly. Based on this, a new kind of support vector machine-μSVM is proposed in the paper. The decision region of the minority class is enlarged by adjusting the distance measurement rule in the classifying decision function. Through theoretical analysis and empirical study, we show that our method augments the classification accuracy rate effectively without increasing the computation complexity.